Igt-dev Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Jesse Zhang <jesse.zhang@amd.com>
To: <igt-dev@lists.freedesktop.org>
Cc: Vitaly Prosyak <vitaly.prosyak@amd.com>,
	Alex Deucher <alexander.deucher@amd.com>,
	Christian Koenig <christian.koenig@amd.com>,
	Kamil Konieczny <kamil.konieczny@linux.intel.com>,
	Jesse Zhang <jesse.zhang@amd.com>
Subject: [PATCH i-g-t] tests/amd_basic: fix eviction test failure
Date: Mon, 25 Mar 2024 14:55:34 +0800	[thread overview]
Message-ID: <20240325065534.2421803-1-jesse.zhang@amd.com> (raw)

 Test case name:igt@amdgpu/amd_basic@eviction-test-with-ip-dma@eviction_test
 Fail error: "Starting subtest: eviction-test-with-IP-DMA
 Starting dynamic subtest: eviction_test
 (amd_basic:9425) amdgpu/amd_command_submission-CRITICAL: Test assertion failure function amdgpu_test_exec_cs_helper, file ../lib/amdgpu/amd_command_submission.c:77:
 (amd_basic:9425) amdgpu/amd_command_submission-CRITICAL: Failed assertion: r == 0
 (amd_basic:9425) amdgpu/amd_command_submission-CRITICAL: Last errno: 12, Cannot allocate memory
 (amd_basic:9425) amdgpu/amd_command_submission-CRITICAL: error: -12 != 0
 Dynamic subtest eviction_test failed.

There is a case where the system size and gtt memory are smaller.
When evicting vram to gtt or system memory, there is not enough gtt or memory system memory available for allocation.
Therefore, try to reduce the use of gtt during initialization to meet the requirements for evicting vram.

 Cc: Vitaly Prosyak <vitaly.prosyak@amd.com>
 Cc: Alex Deucher <alexander.deucher@amd.com>
 Cc: Christian Koenig <christian.koenig@amd.com>
 Cc: Kamil Konieczny <kamil.konieczny@linux.intel.com>

 Signed-off-by: Jesse Zhang <Jesse.Zhang@amd.com>
---
 tests/amdgpu/amd_basic.c | 16 ++++++++++++----
 1 file changed, 12 insertions(+), 4 deletions(-)

diff --git a/tests/amdgpu/amd_basic.c b/tests/amdgpu/amd_basic.c
index 70e45649d..8adb8d696 100644
--- a/tests/amdgpu/amd_basic.c
+++ b/tests/amdgpu/amd_basic.c
@@ -340,6 +340,7 @@ amdgpu_bo_eviction_test(amdgpu_device_handle device_handle)
 	uint64_t gtt_flags[2] = {0, AMDGPU_GEM_CREATE_CPU_GTT_USWC};
 
 	const struct amdgpu_ip_block_version *ip_block = get_ip_block(device_handle, AMDGPU_HW_IP_DMA);
+	uint64_t gtt_allocation;
 
 	igt_assert(ip_block);
 
@@ -358,6 +359,17 @@ amdgpu_bo_eviction_test(amdgpu_device_handle device_handle)
 				   0, &vram_info);
 	igt_assert_eq(r, 0);
 
+	r = amdgpu_query_heap_info(device_handle, AMDGPU_GEM_DOMAIN_GTT,
+				   0, &gtt_info);
+	igt_assert_eq(r, 0);
+
+	/* For smaller gtt memory sizes, reduce gtt usage on initialization
+	 * to satisfy eviction vram requirements */
+	if (gtt_info.heap_size - gtt_info.max_allocation < vram_info.max_allocation)
+		gtt_allocation = gtt_info.max_allocation/3;
+	else
+		gtt_allocation = gtt_info.max_allocation;
+
 	r = amdgpu_bo_alloc_wrap(device_handle, vram_info.max_allocation, 4096,
 				 AMDGPU_GEM_DOMAIN_VRAM, 0, &ring_context->boa_vram[0]);
 	igt_assert_eq(r, 0);
@@ -365,10 +377,6 @@ amdgpu_bo_eviction_test(amdgpu_device_handle device_handle)
 				 AMDGPU_GEM_DOMAIN_VRAM, 0, &ring_context->boa_vram[1]);
 	igt_assert_eq(r, 0);
 
-	r = amdgpu_query_heap_info(device_handle, AMDGPU_GEM_DOMAIN_GTT,
-				   0, &gtt_info);
-	igt_assert_eq(r, 0);
-
 	r = amdgpu_bo_alloc_wrap(device_handle, gtt_info.max_allocation, 4096,
 				 AMDGPU_GEM_DOMAIN_GTT, 0, &ring_context->boa_gtt[0]);
 	igt_assert_eq(r, 0);
-- 
2.25.1


             reply	other threads:[~2024-03-25  6:55 UTC|newest]

Thread overview: 5+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-03-25  6:55 Jesse Zhang [this message]
2024-03-25  7:49 ` ✓ CI.xeBAT: success for tests/amd_basic: fix eviction test failure Patchwork
2024-03-25  7:59 ` ✓ Fi.CI.BAT: " Patchwork
2024-03-25  9:13 ` ✗ Fi.CI.IGT: failure " Patchwork
2024-03-25 15:50 ` [PATCH i-g-t] " Kamil Konieczny

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20240325065534.2421803-1-jesse.zhang@amd.com \
    --to=jesse.zhang@amd.com \
    --cc=alexander.deucher@amd.com \
    --cc=christian.koenig@amd.com \
    --cc=igt-dev@lists.freedesktop.org \
    --cc=kamil.konieczny@linux.intel.com \
    --cc=vitaly.prosyak@amd.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox